Fig. 6: Rate of increase of \(\overline{{{{{\mathscr{E}}}}}_{d}}({{{\mathcal{X}}}})={c}_{d}(J)\gamma t\) as a function of qudit dimension.
From: Noisy qudit vs multiple qubits: conditions on gate efficiency for enhancing fidelity
The data shown was generated for \(H={{\mathbb{0}}}_{d}\) and γt ∈ [0, 10−4]. The markers show the numerical results. The solid curves represent the expected linear responses according to (16). Each marker/colour pair corresponds to a different error channel \({{{\mathcal{X}}}}\), with collapse operators J specified in the legend.
